Origin and significance of the production of carbon dioxide during the ozonization of 13C-labeled D-glucose at different pH values

Article abstract of DOI:10.1016/S0008-6215(01)00068-4

[1-¹³C], [2-¹³C] and [6-¹³C] D-glucose were, respectively, ozonized in a semi-batch reactor in acidic and basic conditions. The composition of the gas phase was evaluated by on-line mass spectrometry measurements. The quantitative and isotopic analyses of the carbon dioxide formed during ozonization are presented and discussed. The data, correlated with previous literature results, clearly show that at pH 2.5 the production of carbon dioxide from C-6 and C-1 carbon atoms is nearly equivalent. Conversely, at higher pH values, CO₂ is released with a greater selectivity from the reducing end. The importance of the decarboxylation reaction in the formation of by-products with fewer than six carbon atoms is also demonstrated.

Full text of DOI:10.1016/S0008-6215(01)00068-4

www.elsevier.com/locate/carres
Carbohydrate Research 333 (2001) 233–240
Origin and significance of the production of carbon
13
dioxide during the ozonization of C-labeled
D-glucose
at different pH valuesꢀ
Olivier Marcq,a Jean-Michel Barbe,a Alain Trichet,b Roger Guilarda,*
aLaboratoire d’Inge´nierie Mole´culaire pour la Se´paration et les Applications des Gaz, LIMSAG UMR 5633,
Faculte´ des Sciences ‘Gabriel’, Uni6ersite´ de Bourgogne 6, Boule6ard Gabriel, F-21100 Dijon, France
bAir Liquide, Centre de Recherche Claude-Delorme, Les Loges en Josas BP 126, F-78350 Jouy en Josas, France
Received 19 December 2000; accepted 28 February 2001
Abstract
[1-13C], [2-13C] and [6-13C]
D-glucose were, respectively, ozonized in a semi-batch reactor in acidic and basic
conditions. The composition of the gas phase was evaluated by on-line mass spectrometry measurements. The
quantitative and isotopic analyses of the carbon dioxide formed during ozonization are presented and discussed. The
data, correlated with previous literature results, clearly show that at pH 2.5 the production of carbon dioxide from
C-6 and C-1 carbon atoms is nearly equivalent. Conversely, at higher pH values, CO2 is released with a greater
selectivity from the reducing end. The importance of the decarboxylation reaction in the formation of by-products
with fewer than six carbon atoms is also demonstrated. © 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Ozone; Glucose oxidation; Carbon dioxide; 13C-labeled glucose; Isotopic analysis
1. Introduction
other conditions.4 The oxidation with oxygen
in alkaline media has been extensively studied
and carbon dioxide formation is also reported,
but as a secondary by-product.5,6
The production of carbon dioxide during
the oxidation of carbohydrates is a well-
known phenomenon which has been demon-
strated in many cases. A famous example is
the Ruff degradation2,3 which leads to ara-
binose by oxidative decarboxylation of cal-
cium gluconate in the presence of hydrogen
peroxide and Fe3+ cations (Scheme 1).
Several studies have shown and often mea-
sured the production of carbon dioxide during
the ozonization of sugars, starch, cellulose,
cellulose model compounds or monosaccha-
rides.7–15 Over the past few years, our labora-
tory tried to improve understanding of the
ozonization reaction pathways of lignin model
The formation of carbon dioxide is also
observed during the oxidation of carbohy-
drates with hydrogen peroxide under many
compounds16 and simple carbohydrates (
-
D
glucose, methyl- -glucopyranose and cel-
D
lobiose)7,8 and evidenced in every case, the
importance of carbon dioxide formation un-
der acidic conditions. At neutral pH, Defaye
and coworkers10 demonstrated the key-role of
the decarboxylation process in mass loss dur-
ing cellulose ozonization.
ꢀ For preliminary communications on production of car-
bon dioxide in acidic conditions, see Ref. 1.
* Corresponding author. Tel.: +33-380-396111; fax: +33-
380-396117.
E-mail address: roger.guilard@u-bourgogne.fr (R. Guilard).
0008-6215/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(01)00068-4

234
O. Marcq et al. / Carbohydrate Research 333 (2001) 233–240
Scheme 1. Mechanism of the Ruff degradation.2,3
with hydrogen peroxide, oxygen, chlorine, or
The origin of carbon dioxide formed during
periodate3,6,22 have appeared in the literature.
The aim of such studies was, in most cases, to
get a better understanding of complicated oxi-
dation reactions and to deal mostly with the
origin of carbon dioxide production. How-
ever, to our knowledge, no ozonization reac-
tion of labeled sugars has ever been carried
out. Therefore, we performed a detailed study
at pH 2.5 in the presence of sulfuric acid or
acetic acid and at pH 10 (sodium hydroxide).
During these reactions, the gas phase was
monitored with an online mass spectrometer
to provide an isotopic and qualitative analysis
of this phase. In order to assess the early
production of carbon dioxide and also to be
able to detect in real time very low levels of
this by-product, a suitable procedure was de-
veloped and is described in Section 2.
ozonization of sugars and especially glucose
or cellobiose has not been clearly established
and no convergence could be found in the
literature. Gluconic acid is generally observed
as the main degradation product8,11–14,17,18 and
Van Nifterik and coworkers14 explained the
formation of carbon dioxide and
during the ozonization of -glucose and cel-
D
-arabinose
D
lobiose by a Ruff type of degradation of glu-
conic acid. A similar pathway was also
proposed by Katai and coworkers12 for the
formation of arabinose. More recently, a Ruff
type degradation was also considered by
Holen and coworkers11,18 but they concluded
that this pathway was probably not the main
source of carbon dioxide. From g-radiolysis
experiments, Schuchman and Von Sonntag15
postulated that carbon dioxide as well as for-
mic acid were formed directly from glucose at
low pH via the decomposition of an hydrotri-
oxide intermediate. In the latter references, the
anomeric position of glucose or cellobiose is
considered as the most reactive and carbon
dioxide is assumed to originate from this
position.
2. Experimental
[1-13C], [2-13C] and [6-13C]
D
-glucose were
purchased from Omicron Biochemical Inc.
(USA). Ozone was generated from pure oxy-
gen (‘C’ quality from Air Liquide Co.) or a
standard mixture of CO2 (0.5 ppm) in pure
oxygen (Air Liquide Co.). Solutions of sugars
were prepared with ultrapure water (18 MV
Elga). The O3–O2 mixture was generated with
a cold-plasma ozonizer (Labo-Lox Trailigaz)
and the concentration of ozone before and
after the semi-batch reactor was monitored
using UV-analyzers (BMT 961) and the flow
rate controlled with a mass flow-meter
(Alphagaz).
For other authors7,8,10,19 the reactivity at
C-6 (primary alcohol) is not negligible com-
pared to that of the anomeric position of
mono- or disaccharides. Rey and coworkers20
even concluded that C-6 was more reactive.
Moreover, some reports on polysaccharides
showed the importance of the oxidation at the
C-6 position9,10,21
.
The importance of the production of carbon
dioxide in such oxidation reactions,7 and our
interest in discovering its source have led us to
investigate the ozonization of specifically la-
beled glucose molecules. A few reports on
oxidation reactions of labeled carbohydrates
Typical ozonization procedure in a semi-
batch reactor.—The substrate was dissolved
(250 mg) in 100 mL of pure water, 1 h before

O. Marcq et al. / Carbohydrate Research 333 (2001) 233–240
235
the reaction, then the tubing and the solution
were purged with the CO2–O2 mixture. The
pH was set with concd H2SO4, glacial AcOH
or 250 mM NaOH solution, 2 min prior to the
introduction of the ozone mixture into the
reactor. Before bubbling into the solution, the
O3–O2 mixture (55–60 mg/L, generated from
the standard mixture of CO2 in O2), was
passed through the tubing and over the void
volume above the solution surface at a flow
rate of 120 mL/min. Once the concentrations
of ozone before and after the reaction flask
were equal, the flow rate was brought to 60
mL/min and the oxidizing mixture was bub-
bled into the solution under magnetic stirring.
The pH was regulated throughout the reac-
tions by automatic titration with a 250 mM
NaOH solution.
The various purges carried out in this pro-
cedure are of major importance. Indeed, the
purge of the connection tubing with the CO2–
O2 enables zero of the analytical instruments
to be set and any exogenous CO2 to be re-
moved from the solution. The purge at high-
flow rate with the oxidizing mixture before
starting the reaction was performed in order
for the zero of ozone consumption to be set.
Quantitati6e analysis of the gas phase.—The
gas phase was continuously analyzed using a
mass-spectrometer (GasLab–Fisons; electron
impact mode) plugged on line at the outlet of
the reactor. The reliability of the measure-
ments was assessed by analyzing ambient air
and the standard mixture (CO2 in O2). The
data acquisition was started before the purge
of the tubing with the oxidizing mixture, so
that the amount of CO2 or CO produced by
the ozonizer, or by reaction of ozone with the
tubing was measured. An average level of
carbon dioxide production during this purge
was computed and subtracted from the values
acquired during the ozonization reaction. In
each case, data corresponding to the molecu-
lar ion M+ (for instance m/z 44 for 12CO2)
’
were added to data corresponding to the ion
13
of mass M+1 (for instance m/z 45 for CO2)
in order to determine the overall production
of a component of the gas phase. An example
of the recorded curves is given in Fig. 1. The
production of CO2 (% vol) is plotted against
reaction time and three zones can clearly be
defined corresponding, respectively, to the
purge with the standard gas, to the purge with
the oxidizing mixture and to the beginning of
the ozonization reaction.
Isotopic analysis of the gas phase.—The ex-
periment was carried out at atmospheric pres-
sure. The mass-spectrometer was used in the
M.I.M. mode (multiple ion monitoring) and
the data were recorded as partial pressures.
The m/z values of the recorded ions were
equal to 16, 18, 28, 29, 30, 32, 44, 45 and 48,
corresponding, respectively, to O, H2O, N2
and/or 12CO, 13CO and/or H12CO, H13CO, O2,
13
12CO2, CO2 and O3. In these experiments, we
determined the contribution of the labeled
carbon atoms in the production of compounds
in the gas phase, especially CO2. Therefore,
for CO2, we expressed the percentage of CO2,
noted %*CO2, containing the labeled carbon
atom from the partial pressures (%) of 13CO2
(i.e., m/z 45) and of 12CO2 (i.e., m/z 44). The
value corresponding to 13CO2 was corrected
for the contribution of 13CO2 derived from
‘natural’ CO2. The same calculation was done
during the ozonization of all the glucose la-
beled isomers to provide values noted
%*C1O2, %*C2O2 and %*C6O2, correspond-
ing, respectively, to the percentage of CO2
originating from C-1, C-2 and C-6. The repro-
ducibility of CO2 formation from one experi-
ment to another was checked to allow
comparisons between the values %*CO2.
Analysis of the liquid phase.—The concen-
tration of glucose in the liquid phase was
analyzed either by gas chromatography–mass
Fig. 1. Variation of the volumic percentage of CO2 in the gas
phase during purge phases and at the beginning of the
ozonization reaction of glucose (pH 2.5, H2SO4).

236
O. Marcq et al. / Carbohydrate Research 333 (2001) 233–240
indicate a good reproducibility between all the
experiments and the reliability of the analyti-
cal techniques.
13C NMR spectroscopy was used to follow
the consumption of glucose as can be seen in
Fig. 2. Samples (400 mL) were added to a
solution (200 mL) of NaH13CO3 used as an
1
internal standard (5 mg/mL). H decoupled
13C NMR experiments with a 30° pulse and
900 accumulations were performed for quan-
tification purposes on a Bruker DRX Avance
500 MHz spectrometer. Such experiments also
allowed the estimation of the concentration of
labeled formic acid formed during the
ozonization of labeled glucose. The signal of
formic acid was observed at 171.4 ppm (CꢀO
resonance).11
Fig. 2. Reproducibility of the ozonization reaction with re-
gards to the consumption of starting material. (") [1-13C]
D-glucose (NMR/Experiment 1); (ꢁ) [1-13C]
D
-glucose
-glucose (HPLC/Experi-
-glucose (GCMS/Experiment 2); (×)
-glucose (NMR/Experiment 1).
(GCMS/Experiment 2); (ꢂ) [2-13C]
ment 1); (*) [2-13C]
[6-13C]
D
D
D
3. Results and discussion
Ozonization reaction at pH 2.5.—Our first
interest was to confirm the importance of the
production of carbon dioxide at pH 2.5 ad-
justed with concentrated sulfuric acid. As
shown in Fig. 3, this result is clearly
evidenced.
Moreover, by using the specific experimen-
tal conditions used here (i.e., purges and use
of a standard mixture) and continuous moni-
toring of the carbon dioxide production, we
observed (Fig. 1) that the production of car-
bon dioxide starts almost immediately after
the introduction of the oxidizing mixture in
the solution. This indicates that carbon diox-
ide is not a final product of the oxidation
process but one of the first in the conditions
described here.
Fig. 3. Influence of the reaction conditions on the production
of carbon dioxide during the ozonization of -glucose. (ꢃ)
pH 2.5 (H2SO4); (--ꢁ--) pH 2.5 (AcOH); (--"--) pH 10
D
(NaOH).
spectrometry (Hewlett–Packard 5890/5972
MSD) after silylation using a high perfor-
mance capillary column HP-1-MS, or by high
pressure anionic exchange chromatography
coupled with conductimetric detection
(HPAEC/CD, Dionex Inc.) using a AS11
column (Dionex Inc.). On the graph, drawn in
Fig. 2, the consumption of glucose is plotted
against the consumption of O3 expressed as
equivalents of the initial substrate. The con-
sumption of O3 is systematically expressed
that way in the other graphs presented in this
article. The curves displayed in Fig. 2 follow
almost exactly the same trend. These data
It has to be noted that the formation of CO
and further oxidation to CO2 in the presence
of ozone is unlikely to happen under the
experimental conditions used.23 At room tem-
perature and atmospheric pressure, the oxida-
tion of CO to CO2 is indeed very slow. No
carbon monoxide has been observed by mass
13
spectrometry (m/z 28, CO and 29, CO) dur-
ing the reaction at pH 2.5 either with sulfuric
or acetic acid (see Section 3.2). Moreover, in
all cases, neither a signal at m/z 30 (parent
peak [H13CO]+ of H13CHO) in the gas phase,
nor HCHO in solution were detected. The

O. Marcq et al. / Carbohydrate Research 333 (2001) 233–240
237
that the contributions of C-1 and C-6 carbon
atoms to the overall production of carbon
dioxide are nearly equal. Even a larger part of
the carbon dioxide may originate from C-6
(47%) rather than from C-1 (36%) carbon
atom between 0.25 and 1 equiv of ozone con-
reproducibility of our experiments at pH 2.5
(with H2SO4) allowed us to use NMR as an
analytical tool to determine the quantity of
formic acid formed. By cumulating the
amounts of formic acid measured by this
method during the ozonization of the three
labeled glucose molecules, we could obtain an
estimation of the overall amount of formic
acid formed from these positions. The amount
of carbon dioxide formed throughout the re-
action is larger than the amount of formic
acid (approx 0.075 mmol and 0.05 mmol,
respectively, after 1 equiv of ozone consumed,
and 0.22 mmol and 0.06 mmol after 2 equiv of
ozone consumed). Carbon dioxide appears as
the only gaseous by-product and as the main
1-carbon atom compound formed at pH 2.5.
The isotopic compositions of CO2 produced
sumed. [2-13C]
D
-glucose has been also
ozonized in sulfuric acid medium. The average
values of %*CO2 between 1 and 3 equiv of
ozone consumed for the ozonization of glu-
cose labeled in C-1, C-2 and C-6 positions are,
respectively, 39, 20 and 37.5%. These values
are only averages but the linear shape of the
curves indicates that the decarboxylation rate
is fairly constant after 0.5 equiv of ozone
consumed which may imply a reproducible
reaction mechanism throughout the reaction.
It appears also that CO2 originates almost
exclusively from carbon atoms in C-1, C-2 and
C-6 positions (more than 90–95% of the over-
all production of carbon dioxide) of the start-
ing glucose molecule.
The two extremities of the glucose molecule
(1 and 6) account for the major part of the
carbon dioxide production which starts
quickly after the beginning of the reaction. We
can hence conclude that the C-6 carbon atom
is affected by the presence of ozone as fast as
the reducing end of the glucose molecule, at
least as far as decarboxylation is concerned in
the presence of sulfuric acid.
from [1-13C], [2-13C] and [6-13C]
D
-glucose,
especially at the beginning of the reactions,
were determined to establish whether it was
the result of random degradations or of pro-
cesses resulting in the loss of C-1, C-2 or C-6
specifically. The percentage of carbon dioxide
derived from each labeled position was com-
puted as described in Section 2. Values of
%C1O2 and %C6O2 (averages of ten measure-
ments) are plotted in Fig. 4 as a function of
ozone consumption.
This figure clearly shows that decarboxyla-
tion reactions at C-1 and C-6 positions can
account for at least 80% of the carbon dioxide
formed during the ozonization of glucose at
pH 2.5 (H2SO4). It has also to be emphasized
Some reports on ozonization of carbohy-
drates have focussed on the use of acetic acid
as a solvent or to set the pH.11,24 For example,
acetic acid has been used as a solvent by
Mbachu25 in bleaching reactions with ozone.
Holen11 has recently investigated the effect of
this acid used to set the pH. These authors
demonstrated that acetic acid has a protective
effect on a cellulose model compound (cel-
lobiose) during ozonization. Holen and
coworkers11,18 also observed an enhanced pro-
duction of gluconic acid during ozonization of
glucose in the presence of acetic acid. It was of
interest for us to investigate the influence of
acetic acid on the production of carbon diox-
ide and to compare the results with those
obtained in the presence of sulfuric acid. We
ozonized [1-13C] and [6-13C]
-glucose in acetic
D
Fig. 4. Evolution of the percentages of carbon dioxide con-
taining, respectively, C-1 (*C1O2) and C-6 (*C6O2) carbon
atoms during the ozonization of glucose at pH 2.5 (H2SO4).
acid, and as Holen and coworkers11,18 had
found, no significant difference was noted be-

238
O. Marcq et al. / Carbohydrate Research 333 (2001) 233–240
over, we can affirm that the same type of
reaction must be considered for the oxidation
of C-6 alcohol function.
Defaye et al.9,10 proposed a stepwise process
from experiments in neutral (pH 7) and
slightly acidic conditions (pH 5). This process,
depicted in Scheme 2, implies the formation of
a carboxylic acid followed by oxidative decar-
boxylation via an a keto acid and occurs at
both ends of the sugar molecule in such a way
that aldaric and a-keto aldaric acids could be
involved.
Our results clearly indicate that such a pro-
cess is likely to happen even if a radical de-
composition of an hydrotrioxyde at C-1
cannot be excluded.8,15 In acidic conditions,
this process would essentially affect C-1 and
C-6 positions and can be extended to account
for the liberation of C-2 as CO2. The produc-
tion of carbon dioxide containing the C-2
carbon atom can be explained by oxidation of
arabinonic acid that can be formed by decar-
Fig. 5. Evolution of the percentages of carbon dioxide con-
taining, respectively, C-1 (13C1O2) and C-6 (13C6O2) carbon
atoms during the ozonization of glucose at pH 2.5 (AcOH).
tween the sulfuric acid and the acetic acid
media with respect to the initial substrate (glu-
cose) consumption, but the formation of by-
products with fewer than six carbon atoms
was less important in the case of acetic acid.26
We also observed that the consumption of
ozone as a function of time followed the same
trend in both acidic media.26 However, as
shown in Fig. 3, the production of carbon
dioxide is much lower during ozonization in
the presence of acetic acid (half the amount of
carbon dioxide produced in the presence of
sulfuric acid after 1 equiv of ozone con-
sumed). The limitation of carbon dioxide for-
mation is hence another aspect of the
protecting effect of acetic acid on glucose.
This is in good agreement with an enhanced
stability of gluconic acid11,18 if we assume that
gluconic acid undergoes a Ruff type of reac-
tion in this medium as well.
boxylation of
D
-arabino-hex-2-ulosonic acid
(2-ketogluconic acid) as we suggested8 in
agreement with the general proposal of a step-
wise mechanism9,10 and by further oxidation
of arabinose.
Ozonization reaction at pH 10.—The experi-
mental conditions developed (see above) also
enabled us to monitor the production of car-
bon dioxide at pH 10 using sodium hydroxide.
The purge of the solution with the standard
mixture of CO2 in oxygen before the reaction
ensured a pre-saturation of the solution so
that once the reaction was started, at least a
part of the carbon dioxide formed evolved fast
enough from the solution to be analyzed by
mass-spectrometry. The reaction at pH 10 was
The plots of %*C1O2 and %*C6O2 values
against consumption of ozone are given in
Fig. 5. Carbon dioxide originating from C-6 is
obviously observed and, it can therefore be
concluded that C-6 is also oxidized by ozone
in the presence of acetic acid.
From the results obtained under acidic con-
ditions, the decarboxylation reaction appears
clearly as a key step in the glucose oxidation
process after the oxidation of the reducing
end. The occurrence of a reaction at the
anomeric position followed by a reaction
analog to the Ruff degradation is well sup-
ported by the analysis of the gas phase. More-
conducted only on [1-13C]
D
-glucose and it
should be noted that in these conditions again,
neither carbon monoxide (determined by
mass-spectrometry), nor labeled formic acid
(determined by NMR) were observed. The
production of carbon dioxide hence is low but
not negligible as represented in Fig. 3.
The very limited production of compounds
with only one carbon atom in alkaline condi-
tions is in agreement with results reported by
our laboratory.7,8 However, we showed that
some carbon dioxide is formed implying that
decarboxylation is still involved in this oxida-

O. Marcq et al. / Carbohydrate Research 333 (2001) 233–240
239
Scheme 2. Decarboxylation process via a-keto-carboxylic acids.
tion of glucose. The isotopic analysis of the
evolved carbon dioxide again allowed a link to
be made between CO2 production and the
formation of product derived by loss of C-1.
The specificity of the ozonization in alkaline
conditions (formation of carbonates) pre-
vented us from determining the amount of
carbon dioxide produced by the ozonizer as
accurately as under low pH conditions. As a
consequence, the production of carbon diox-
ide being low at the beginning of the reaction
and the participation of C-1 carbon being very
high, the calculated percentages %*C1O2 were
found to be even greater than 100%. Never-
theless the information displayed in Fig. 6 is
consistent with more selective oxidation at the
anomeric position and the formation of a
carboxylic group at this position.7,8
The formation of carbon dioxide exclusively
from the C-1 position at the beginning of the
reaction under alkaline conditions strongly
supports the mechanism proposed in Scheme
2. The decarboxylation of arabinonic acid to
produce carbon dioxide containing the C-2
carbon atom could be related to the decrease
of the percentage of carbon dioxide containing
the C-1 carbon atom after 1 equiv of ozone
consumed (Fig. 6).
4. Conclusion
It is obvious that with glucose the decar-
boxylation is not a random reaction but is a
key-step in the formation of by-products with
fewer than six carbon atoms. The early forma-
tion of carbon dioxide indicates that this com-
pound is not the result of the decomposition
of secondary by-products. The detailed study
carried out at pH 2.5 (sulfuric acid or acetic
acid) leads to the conclusion that oxidations at
C-6 and C-1 contribute similarly to the forma-
tion of products with fewer than six carbon
atoms. Conversely, at high pH values, CO2 is
released with greater selectivity from the re-
ducing end.
In all cases, correlations with literature and
our results support some previous proposals
and the limitation of the decarboxylation reac-
tion has been illustrated to be another exam-
ple of the ‘protective effect’ of acetic acid.
Our results clearly show that the formation
of carbon dioxide should therefore be investi-
gated in any application of ozone involving
oxidation of carbohydrates. Reaction path-
ways will be further discussed along with the
quantitative and isotopic analysis of the liquid
phase in a forthcoming paper.†
Acknowledgements
The ‘Centre National de la Recherche Sci-
entifique’ (C.N.R.S.), Air Liquide Co. and the
‘Conseil Re´gional de Bourgogne’ are grate-
fully acknowledged for the financial support
to this study. O.M. thanks the CNRS and the
Fig. 6. Evolution of the percentage of carbon dioxide contain-
ing C-1 (*C1O2) carbon atom during the ozonization of
glucose at pH 10 (NaOH).
† For preliminary communications, see Ref. 26.

240
O. Marcq et al. / Carbohydrate Research 333 (2001) 233–240
13. De Laat, J.; Bawa, M.; Dore, M. En6iron. Technol. 1995,
16, 1035–1048.
14. (a) Van Nifterik, L.; Xu, J.; Rakoto, C.; Laurent, J. L.;
Mathieu, J.; Molinier, J.; Coste, C.; Kalck, P. Second
European Workshop on lignocellulosics and Pulp, Greno-
ble, France, 1992, pp. 165–166;
‘Conseil Re´gional de Bourgogne’ for a Ph.D.
grant (B.D.I.).
(b) Van Nifterik, L.; Xu, J.; Laurent, J. L.; Mathieu, J. J.
Chromatogr., Sect. A 1993, 640, 335–343;
(c) Van Nifterik, L. Ph.D. Thesis, Universite´ de Tou-
louse, France, 1993.
References
1. (a) Marcq, O.; Barbe, J. -M.; Trichet, A.; Guilard, R.
Glupor3; Aveiro-Portugal, 1999; Abstract of papers, OC-
12-p. 78;
15. Schuchmann, M. N.; von Sonntag, C. J. Water Sci. Res.
Technol. 1989, 38, 311–317.
(b) Marcq, O.; Barbe, J. -M.; Trichet, A.; Guilard, R.
Proceedings of the International Ozone Symposium, IOA,
Basel, Switzerland, 1999, pp. 183–185.
2. Ruff, O.; Ollendorf, G. Ber. Dtsch. Chem. Ges. 1900, 31,
1573–1577.
3. Isbell, H. S.; Salam, M. A. J. Carbohydr. Chem. 1981, 88,
123–126.
16. Le´tumier, F. Ph.D. Thesis, Universite´ de Bourgogne,
France, 1997.
17. (a) Pan, G. Y. Y.; Chen, C.-L.; Chang, H.-M.; Gratzl, J.
S. J. Wood Chem. Technol. 1984, 4, 367–387;
(b) Pan, G. Y. Y.; Chen, C.-L.; Gratzl, J. S.; Chang,
H.-M. Res. Chem. Intermed. 1995, 21, 205–222;
(c) Bawa, M. L. Ph.D. Thesis, Universite´ de Poitiers,
France, 1992.
18. (a) Holen, A. K.; Moe, S. T.; Christensen, P. K. Fourth
European Workshop on Cellulosics and Pulp, Stresa, Italy,
1996, pp. 475–480;
4. (a) Haskins, J. F.; Hogsed, M. J. J. Org. Chem. 1950, 15,
1264–1274;
(b) Moody, G. J. Ad6. Carbohydr. Chem. 1964, 19, 174–
179.
5. (a) Malinen, R.; Sjo¨stro¨m, E. Pap. Puu-Pap. Timb. 1972,
54, 451–468;
(b) Holen, A. K.; Kleppe, P. J.; Moe, S. T. International
Symposium on Wood and Pulp Chemistry, Book 2,
Helsinki, Finland, 1998, pp. 481–486.
(b) Warshowsky, B.; Sandstrom, W. M. Arch. Biochem.
Biophys. 1952, 37, 46–52;
(c) Malinen, R.; Sjo¨stro¨m, E.; Ylijoki, J. Pap. Puu-Pap.
Timb. 1973, 55, 5–13;
(d) Malinen, R.; Sjo¨stro¨m, E. Pap. Puu-Pap. Timb. 1973,
55, 547–556;
(e) Ericsson, B.; Lindgren, B. O.; Theander, O. Cellulose
Chem. Technol. 1973, 7, 581–591.
19. (a) Kishimoto, T.; Nakatsubo, F.; Murakami, K.;
Umezawa, T. J. Wood Chem. Technol. 1995, 15, 453–471;
(b) Kishimoto, T.; Nakatsubo, F. Holzforschung 1996, 50,
372–378.
20. Rey, R. P.; Moraleda, M. G.; Lazcano, R. R. Eight
Ozone World Congress, Proceedings section E, Zurich,
Switzerland, 1987, pp. 106–127.
6. (a) Phillips, G. O.; Rickards, T. J. Chem. Soc. B 1969,
455–461;
(b) Kratzl, V. K.; Schwarz, H. A. Holzforschung 1975, 29,
29–31.
7. Lemeune, S. Ph.D. Thesis, Universite´ de Bourgogne,
France, 1996.
21. Szymanski, C. D. J. Appl. Polym. Sci. 1964, 8, 1597–
1606.
22. (a) Daniel, J. W. Tappi 1959, 42, 534–545;
(b) Crossman, J. K.; Green, J. W. Tappi 1966, 49, 131–
136;
(c) Eisenberg, F. J. Am. Chem. Soc. 1954, 76, 5152–5154;
(d) Isbell, H. S.; Sniegoski, L. T. J. Res. NBS, Sect A
1964, 68, 301–304.
8. Lemeune, S.; Barbe, J.-M.; Trichet, A.; Guilard, R.
Ozone Sci. Eng. 2000, 22, 447–460.
9. Gadelle, A. Ph.D. Thesis, Universite´ de Grenoble,
France, 1984.
10. Angibeaud, P.; Defaye, J.; Gadelle, A. In Cellulose and its
Deri6ati6es; Kennedy, J. F., Ed.; Ellis Horwood: New
York, 1985; pp. 161–171.
11. Holen, A. K. Ph.D. Thesis, Norwegian University of
Science and Technology, Norway, 1998.
12. (a) Katai, A. A. Ph.D. Thesis, Syracuse University, Syra-
cuse, NY, 1965;
23. (a) Hartek, P.; Dondes, S. J. Chem. Phys. 1957, 26,
1734–1737;
(b) Arin, L. M.; Warneck, P. J. Chem. Phys. 1972, 76,
1514–1516.
24. Kesselmans, R. P. W.; Bleeker, I. P. EP 0885245A/WO
97/35890, 1998.
25. Mbachu, R. A. D.; St John Manley, R. Tappi 1981, 61,
67–70.
26. Marcq, O.; Barbe, J. -M.; Trichet, A.; Guilard, R. 220th
National Meeting of the American Chemical Society,
Washington DC, USA, 2000, abstr. Carb-61.
(b) Katai, A. A.; Schuerch, C. J. Polym. Sci. Polym.
Chem. 1966, 4, 2683–2703.
.